(1) Godot's renderer is technical ass-- it can make a pretty scene, but it does not scale well to games. FPS drops, stitching, and more artifacting than every Indiana Jones and Lara Croft movie/game combined.
(2) The WHOLE engine is hideously unoptimized-- 5 years ago: https://github.com/godotengine/godot/issues/23998 ... still a problem today. The engine itself is a bottleneck to any performance. Also, this recently... https://sampruden.github.io/posts/godot-is-not-the-new-unity/ ... I wasn't aware of how bad this actually was, as I didn't use C# in Godot. Godot, itself, is a bottleneck to anything performant.
Another AAA engineer took a technical look through Godot's source code: https://blog.odorchaidhe.games/posts/godot/ They have come to the same conclusion I did years ago. How many /actual/ pros need to tell you your engine is not for large games before you actually /listen/?
(3) Asset importing puts the ass in assets-- good luck importing anything more than the simplest animated assets into Godot. If you get lucky, you might... but, then good luck actually loading larger PBR scenes in Godot. Demo scenes, sure... but actual full on game levels? The team I worked with had to move to Unreal because Godot couldn't load a level with any serious fidelity (well, just ONE of the reasons).
(4) Built by hobbyists, FOR hobbyists. The core philosophy of Godot is to build for newbies... you can't be an engine that wants its source code readible by newbies and have optimzied code at the same time. Those two things are very anti-thetical of each other. Godot is a great game jam engine... and, if you have smaller games... you can use it to build some commercial games. If you look at every single commercial hit in Godot... they are all technically small games. But this is the most important part: GODOT DOES NOT SCALE. As your node numbers climb, engine performance drops significantly. If you can actually manage to get Godot to load a larger game level and run it... good luck running it on anything but cutting edge systems. People often forget that their pretty demos won't run on machines even a few years old. People say "Nuh huh, Sonic Colors used it"... yeah, and if you catch them in private in an honest moment they will tell you they absolutely regretted using it.
(5) Godot is not community driven as they like to say it is-- it is 100% Juan driven. Juan does what Juan wants... and Juan doesn't do what Juan don't Juan-na. Including adding feaures engines need, fixing performance issues, etc. Godot suffers from "I'll do it myself later" syndrome. The "leader" of Godot famously couldn't understand why someone would want a terrain engine for a 3D game because you couldn't make it to fit ALL game use cases... and then followed up by saying "we can never know what terrain tools would be needed". He eventually relented to the possibility of adding terrain... but it took YEARS. The guy has zero experience with 3D tools... and doesn't know his head from his feet. No engine is ever going to do well with that kind of obtuse leadership. Not to mention, this is the same guy who said, "Linked lists are the most efficient way to manage memory." You about ready to face palm, because it gets even better.
(6) Look at the state of Godot 4. That fiasco started in 2018... we said it was going to be fiaso, we told them (various Godot mods even) told them it was going to be a fiasco... and as we tried the alpha we told them it was going to be a disaster. And lo' and behold... a disaster it was. We're nearly at 4.2 and the engine is neither stable nor production ready. Which again, is a throw back to point 4... it's an engine built by hobbyists. It is not a professional team of engineers building Godot, so you will /never/ get another Unity out of Godot.
(7) Five years ago the creator of Rimworld look at using Godot to make games... his conclusion was that Godot is unsuitable for serious game developement because it doesn't address or provide for serious game developers. And he said, and I paraphrase, "In 5 years Godot will just be spinning its tires in the mud and going nowhere". I said the exact same thing in 2018... we were both dead on the money. For reference, the post is here, you can scroll down where Tyrian chimes in: https://www.reddit.com/r/godot/comments/8mhzfo/tynansylvester_of_rimworld_fame_is_evaluating/
(8) Godot constantly adds and leaves features unfinished-- which is why Godot 4 is the shit show it currently is. They keep adding bulk and never fixing it... and not to the degree Unity or Unreal does, but signfiicantly worse. When your engine is neither stable nor production ready a year or two after release... says everything.
(9) Ignore Godot "Tutorial Makers" and their HYPE. None of them make games for a living. Their whole purpose is to get your eyes, your views, and earn money from your hopes and dreams-- they don't give a shit about you or your game or whether or not you succeed, they just want your clicks. None of them have built any significant games to prove what "Godot can do"... because Godot can't do it, period. I've been in the Godot ecosphere for nearly a decade now... and time and time again I have asked people who countered my points to "Show me your game". In all this time, I've yet to be shown a game. Or maybe it's just coincidence alllllll the people who said it can do it just haven't done it. But, I know plenty of people who have tried... and all have moved to other engines for serious 3D games, including myself.
(10) BUT IT IS OPEN SOURCE, YOU CAN FIX IT YOURSELF... oh, can I? So, I can give up working on games to fix every single problem Godot has? Good freakin' luck, guys. That's a LOT of growing problems to deal with. Also, are you a game engine engineer? Can you squeeze Unity or Unreal performance out of Godot? You gonna rewrite the whole core of the engine to make it a powerhouse? If you believe you can, you should be building your own engine... not wasting your time in Godot. Most of us want to build games, NOT engines. It's why we have game engines in the first place, to do the grunt work... but Godot ain't much of a grunt. It's more like a couch sittin' keyboard warrior that yells how good it is but has never even been in a fist fight, let alone seen the blood of combat.
(11) I was a community mod for Godot's discord for a few years. I spent hours and hours of my day, every single day, directly talking to new Godot users all the time from all walks of life-- this often included professional devs from studios who were evaluating Godot for larger projects. There were many times Godot was being evaluated by studios and found lacking-- and they had questions about us about PRs and how long it seemed to get PRs addressed or how they had a back and forth with Juan that left a bad taste in their mouth. Myself and other voice mods tried repeatedly... and I mean repeatedly... for years to pass the concerns of what we were hearing from these people to Godot leadership and they would, essentially, put their fingers in their ears and pretty much go "La la la la la we're not listening". THAT is Godot in a nutshell. Time and time again we were told "things are changing" "things will change"... and things /never/ changed, ever. And they still haven't changed... not one little bit. I quit being a mod the same day Remi told me and I quote "Juan doesn't care about the community, it is his engine". If that's the people you want to put the future of your career in... be my guest, and may godspeed.
So, no... Godot is not going to be the next Unity.
It doesn't have the engineering team, it doesn't have the direction, and even if it had the funding to have all that, even worse... it has Juan, who doesn't know what the hell he is doing as game engine lead and 3D engine developer.
Anyone telling you Godot is going to be the next big thing, especially in 3D... ask them to pony up and show you where their 3D game is that isn't some low poly retro FPS... because I guarantee you, they don't have one... and if they do, it's just a pretty single room or empty field with barely anything in it.
And don't get me wrong here-- I don't hate Godot. I love that scrappy little engine... I use it for small casual games, but it is by no means and measure a "professional grade" engine that usurp something like Unity, no matter how much Unity messes up. Because going from Unity to Godot is like going from a sportscar that occasionally needs some maintenance to riding a tricycle with three flat tires and a broken seat and note saying "fix it yourself".